Identification of novel antifungal agents: antimicrobial evaluation, SAR, ADME–Tox and molecular docking studies of a series of imidazole derivatives

Thirty-four imidazole-based compounds synthesized by one-pot catalytic method were evaluated for their antifungal and antibacterial activities against several fungal and bacterial strains. None of the compounds had antibacterial activity. Interestingly, compounds 1, 2, 3, 10 and 15 displayed a strong antifungal activity against all the tested fungal species, while compounds 5, 7, 9, 11, 21 and 27 showed a moderate antifungal activity. To better understand the biological activity of the most active compounds ADME–Tox and molecular docking studies were carried out. Interestingly, compounds 1, 2, 3, 7, 10 and 15 showed excellent bioavailability. In addition, compounds 1, 2 and 3, exhibited good toxicity profiles. Docking studies of the two most active compounds 2 (IC50 of 95 ± 7.07 μM) and 10 (IC50 of 235 ± 7.07 μM) suggested that they might act by inhibiting the fungal lanosterol 14α-demethylase. Therefore, these novel antifungal agents merit further characterization for the development of new antifungal therapeutics. Electronic supplementary material The online version of this article (10.1186/s13065-019-0623-6) contains supplementary material, which is available to authorized users.


Introduction
Human infectious diseases are among the top ten causes of death worldwide, according to the World Health Organization [1]. Therefore, infectious diseases are among the most important public health problems around the world, especially in developing countries [2]. Drug-resistant infections are also a growing public health threat worldwide [3][4][5]. As a matter of fact, the increasing number of antifungal drug resistance has become more frequent and serious, creating a need for safer and more effective antifungal therapies [6][7][8][9]. Indeed, increasing resistance of fungal pathogens to current antifungal drugs is one of the reasons for the difficulty to tackle fungal infections, particularly in immune-compromised individuals [10]. Given this difficulty in controlling fungal infections, a new class of antifungal drugs with novel mechanisms of action and broad spectrum of activity must be discovered.
In this view, many heterocyclic compounds have been studied in order to discover novel antimicrobial agents. In fact, heterocyclic chemistry has grown considerably and more than 90% of new drugs contain heterocycles [11]. More particularly, imidazole-based heterocyclic compounds occupy a prominent place in heterocyclic chemistry [12,13]. Members of this family of compounds have diverse biological, pharmacological, environmental and industrial applications [14]. Indeed, this type of compounds act as inhibitors of p38 MAP kinase and B-Raf [15], and showed anti-inflammatory, anti-cancer, antifungal, anti-tuberculosis, and anti-diabetic activities [16][17][18]. Accordingly, several imidazole-based heterocyclic compounds have been clinically used to treat many diseases such as Bifonazole which is a clinically used antifungal Open Access agent with a broad spectrum of activity [19,20], Metronidazole used to treat a wide variety of bacterial and parasitic infections [21], Cimetidine used in the treatment of duodenal and gastric ulcers [22], and Eprosartan used as angiotensin II inhibitor and antihypertensive agent [23].
In this context, we sought to study the antifungal and antibacterial biological activities of a series of 34 imidazole-based heterocyclic compounds that have been synthesized by a simple one-pot catalytic method. We have thus identified five imidazole derivatives 1, 2, 3, 10 and 15, which act specifically as potent antifungal agents and lack antibacterial activity. Their structure-activity relationship analysis (SAR), ADME-Tox profiles, and molecular docking studies were carried out to understand their biological activity. These novel antifungal agents merit further characterization and can serve as a promising lead compounds for the discovery of new antifungal therapeutics.

Results and discussion
The imidazole-based compounds studied in this paper were synthesized by one-pot catalytic method [24] and are shown in Fig. 1. The spectroscopic information of these compounds is described below in materials and methods.
Regarding the antibacterial activity, all the compounds were tested for toxicity against three gram-negative bacterial strains (E. coli, C. freundii and S. braenderup) and two gram-positive bacterial strains (L. monocytogenes and S. aureus). All the compounds showed no antibacterial activity against all bacterial strains used. Therefore, to better understand the antifungal activity of the most active compounds, SAR, ADME-Tox and molecular docking studies were carried out and are detailed below.

SAR of imidazole derivatives
The SAR analysis of the mono-arylated series (1-7) revealed that the antifungal activity of these compounds depends essentially on the attached group (R) at the position 1 of imidazole moiety and the attached group (R') of the phenyl ring. Investigation of R moiety in compounds 2, 3, 5 and 7, showed that the presence of OMeC 6 H 4 substituent in compound 5 or CHOC 6 H 4 group in compound 7 resulted in moderate antifungal activity. Whereas, the presence of benzyl substituent in compound 2 or n-Butyl group in compound 3 resulted in high antifungal activity. Further evaluation of compounds 2 and 3 demonstrated that 2 was more potent than 3, with a half-maximal inhibitory concentration (IC 50 ) against S. cerevisiae of 95 ± 7.07 μM and 220 ± 14.14 μM, respectively (Table 1).
These findings suggest that the small size and the electron withdrawing character of the substituent at the position 1 of imidazole moiety are important for the antifungal activity of these compounds. Regarding the comparison between compounds 1 and 2, we observed a stronger antifungal activity against S. cerevisiae with 2 (IC 50 of 95 ± 7.07 μM) compared to 1 (IC 50 of 240 ± 14.14 μM), suggesting that the replacement of chlorine (an electron withdrawing group) with methyl (an electron donating group) decreases the antifungal potential. Together, these data suggest that the introduction of electron withdrawing substituents in R and R' increases the antifungal activity of these compounds.
The SAR analysis of the 2, 5-diarylated imidazole derivatives series (8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26) revealed that the introduction of electron withdrawing or electron donating groups into the phenyl rings at the para position resulted in moderate antifungal activity (9,21) or no activity (12, 14, 17, 18, 20, 22 and 23). We also observed that the presence of electron withdrawing groups [CF 3 (16), CN (13) or NO 2 (19)] into the phenyl rings at the meta position resulted in loss of antifungal activity. However, the presence of an electron donating group (methyl group) at the meta position of the phenyl rings (10) resulted in strong antifungal activity. We also found that the substitution of the phenyl rings at the ortho position with Fluor or methyl groups led to strong antifungal activity (15) or moderate activity (11). Further evaluation of the most active compounds of this series (10 and 15) has demonstrated that compound 10 was more potent than compound 15, with IC 50 's against S. cerevisiae of 235 ± 7.07 μM and 305 ± 21.21 μM, respectively (Table 1). Together, these findings suggest that the antifungal activity of these diarylated imidazole derivatives depends on the size (small or bulky), position (para, ortho or meta) and electronic effect (withdrawing or donating) of the substituent at the phenyl rings.
The SAR studies of the imidazole derivatives series (27 to 34) revealed that the presence of many strong electron withdrawing CF 3 groups at the meta position of phenyl groups did not lead to any antifungal activity. Similarly, the introduction of bulky groups at the position 1 of imidazole moiety did not lead to significant growth inhibition against the tested fungal strains. These findings are consistent with the above observations which suggest that the presence of bulky substituent at the position 1 of imidazole moiety, or at the meta position of phenyl group is unfavorable for the antifungal activity.

ADME-Tox predictions
Currently, the computational predictions of pharmacokinetic and pharmacodynamic parameters such as absorption, distribution, metabolism excretion (ADME) and toxicity risks (Tox) are of great importance in the drug Fig. 2 Antifungal activity of the studied compounds against Saccharomyces cerevisiae, Candida albicans and Candida krusei. Cells were cultured in the presence of 500 µM of each compound for 24 h and growth rate was then assayed by the OD 600 . Growth in the presence of compound was expressed as a percentage relative to the untreated control. All experiments were carried out in triplicate and means were calculated ± SD. *p < 0.05 versus untreated control (See figure on next page.) discovery process [25]. Therefore, in this study we used Molinspiration and DataWarroir programs, as described in Materials and Methods, to determine the ADME-Tox profiles of the most active compounds (1, 2, 3, 7, 10, 11 and 15) as well as the reference antifungal drug Fluconazole. Table 2, in silico prediction of toxicity properties (Mutagenic; Tumorigenic; Irritant; Reproductive effect) of our imidazole derivatives revealed that compounds 1, 2 and 3 have a good toxicological profile with no risk of mutagenicity, tumorigenicity, irritation or reproduction. These Tox properties are similar to those of Fluconazole. However, compound 7 showed a high toxic effect on reproduction probably due to the presence of the C=O carbonyl group at the aldehyde moiety. Regarding 2, 5-diarylated imidazole derivatives 10, 11 and 15, toxicity prediction revealed that these compounds have a low mutagenic effect, but without risk of tumorigenicity, irritation or reproduction. Overall, these Tox studies suggest that compounds 1, 2 and 3 do not have any undesirable moieties involved in toxicity problems, while compounds 10, 11 and 15 exhibit acceptable toxicity profiles with a low mutagenic effect.

As shown in
As stated above, Molinspiration was used to determine pharmacokinetic parameters of the most active compounds. Therefore, the following ADME characteristics were calculated: n-octanol/water partition coefficient characterizing lipophilicity (LogP), molecular weight expressed in Daltons (MW), number of hydrogen bond acceptors (nOH), number of hydrogen bond donors (nOHNH), number of rotatable bonds (nrotb), and total polar surface area (TPSA). As shown in Table 2, all the studied compounds exhibit ADME characteristics that are in agreement with Lipinski's rule of five, which evaluate the drug-likeness, absorption and intestinal permeability of a compound [26,27]. Indeed, for all the compounds LogP are less than 5.0, MW are less than 500, nOH are less than 10, nOHNH are less than 5, nrotb are less than 10 and TPSA are less than 140 Å 2 . Together, these data suggest that these studied  compounds present good bioavailability and therefore can be qualified as a good lead.

Molecular docking studies
Fungal lanosterol 14α-demethylase (CYP51) is an attractive therapeutic target for the development of antifungal drugs [28,29]. This enzyme catalyzes an essential step in the synthesis of ergosterol which is an essential component of fungal cell membrane. CYP51 is the target of azoles, the most popular class of antifungal drugs. There has been a considerable amount of research interest into this enzyme and the azoles because of the dramatically increasing number of drug resistance among certain fungal species [30]. Ketoconazole is an antifungal imidazole that belongs to the azole class and it is currently used to treat a wide variety of fungal infections. Like other azoles, it acts by inhibiting selectively [31]. However, the use of Ketoconazole has been limited because it is associated with clinically important toxic side effects [32,33]. Therefore, it is important to discover novel antifungal imidazoles which act by inhibiting selectively CYP51 without toxic side effects.
Considering the fact that our compounds have imidazole moiety we sought to study the possible binding of our most potent compounds to CYP51 as a possible target protein. Therefore, molecular docking studies were conducted in order to explore the affinity of our most potent imidazole derivatives towards CYP51 from S. cerevisiae. Molecular docking is a very powerful computational method for predicting and modeling the interactions between a small molecule and a protein target at the atomic level. Indeed, molecular docking is an important tool, which is widely used in drug design [34,35]. Table 3 and Fig. 3 show the docking analysis results of compound 2 (the most potent compounds of the mono-arylated series), compound 10 (the most potent compounds of the 2, 5-diarylated imidazole derivatives series), and the antifungal reference drug fluconazole towards CYP51 from S. cerevisiae.
Three parameters were used for docking analysis: the binding affinity expressed in kcal/mol, the interactions between ligand atoms and amino acid residues of the target protein, and the distance of these interactions. Therefore, our results of docking showed that the two tested imidazole derivatives (2 and 10) and fluconazole perform a hydrogen bond interaction (H-bond) between one of  To affirm correlation between in silico prediction of binding affinity to CYP51 and antifungal activity, we also docked two imidazole derivatives which lack antifungal activity (6 and 30) against CYP51 (Table 3). Interestingly, our docking studies revealed that compound 30 does not have binding affinity toward CYP51. Regarding compound 6, the results of docking indicated that this compound have lower binding affinity (− 5.616 kcal/mol) toward CYP51 as compared to the active compounds 2 and 10 (− 6.857 and − 6.791 kcal/mol). Moreover, the docking result revealed that, like compound 30, compound 6 does not interact with CYP51 by hydrogen bond interaction. Together, these docking studies agreed with the antifungal activity and suggest that the two most active compounds 2 and 10 have good binding affinity with target enzyme CYP51 and therefore might act by inhibiting the fungal lanosterol 14α-demethylase.

Conclusion
In this paper we investigated the antifungal and antibacterial biological activities of thirty-four imidazole-based compounds synthesized by one-pot catalytic method. Antifungal activity was assayed against five fungal species, while antibacterial activity was tested against five bacterial species. None of the tested compounds showed an antibacterial activity. Interestingly, compounds 1, 2, 3, 10 and 15 displayed a strong antifungal activity against all the tested fungal species, while compounds 5, 7, 9, 11, 21 and 27 showed a moderate antifungal activity. SAR studies revealed that the antifungal activity of these imidazole derivatives depends on the size (small or bulky), position (para, ortho or meta) and electronic effect (withdrawing or donating) of the substituents at the phenyl rings, as well as the bulkiness and the electronic effect of the substituent in position 1 of the imidazole moiety. ADME analysis showed that compounds 1, 2, 3, 7, 10 and 15 have excellent bioavailability. In addition, Tox studies showed that compounds 1, 2 and 3, have good toxicity profiles, whereas 10, 11 and 15 have low mutagenic effect. On the other hand, compound 7 is likely to cause toxicity because of the high toxic effect on reproduction.
Docking studies of the two most active compounds 2 and 10 suggested that they might act by inhibiting the fungal lanosterol 14α-demethylase. Therefore, these novel antifungal agents merit further characterization and can serve as promising lead compounds for the discovery of new antifungal therapeutics.

Chemistry
The 34 imidazole-based molecules analyzed in this paper have been synthesized by a simple one-pot catalytic method [24]. The spectroscopic information of these compounds is as follows: